2. Limited Dependent Variable Models (60 points): Empirical Model: Loan Default = Bo + BiAge + BzEducation + B3Sex + BASmoke + & Female and Smoke are binary indicator variables ( 0 and 1). Age and Education is measured continuously in years. Loan Default is a binary indicator variable (0 = no default, 1=default). A. Are the predictor variables in Models A statistically significant at the 5% significance level? B. Carefully interpret the coefficients for Female and Education in Models A C. Generate a forecast from all 3 models when Age=10, Educ=10, Sex=1, Smoke=0 D. You are unsure if education has a linear effect on loan default. What are possible transformations you can do to check for non-linearity? Regression Model A: OLS (from R) Coefficients : Estimate Std. Error t value Pr(>It!) (Intercept) 0 . 136269 0 . 011681 11. 666